home › event - data: making it be there when you want it and making it go away when you want it gone
Data: Making it be there when you want it and making it go away when you want it gone
series: Future of Networking
2 December 2010
George E. Pake Auditorium, PARC
Many types of data (e.g., health records), need to be robustly stored for some time, after which the data should be completely destroyed. With traditional systems, making data highly recoverable requires making multiple copies, and then it becomes difficult to assure, when the data is intended to be discarded, that all copies have been destroyed.
This talk describes a file system design that allows setting expiration dates on data. Although all the state in the file system is reliably backed up, once a file expires, it is unrecoverable, even though the backups still exist. Obviously, the data is encrypted, and then the keys are deleted. However, storage of the keys has the same problem; that they need to be backed up, and it will be difficult to assure all copies of a deleted key are destroyed. This file system design solves this problem in a way that is easy to manage and has only minimal overhead compared to a traditional encrypted file system.
Perlman is a Fellow at Intel Labs. Her work on network protocols has had a deep influence on how data moves through the Internet. She designed the spanning tree algorithm which is the heart of how modern Ethernet works, and also made link state routing robust, scalable, and easy to manage. The specific link state protocol she designed for DECnet in the 1980s was standardized by ISO, renamed IS-IS, and remains widely deployed today. Perlman designed TRILL, a new IETF standard intended to replace spanning tree bridging. She is the author of two well-known textbooks: “Interconnections” and “Network Security,” and has received many industry awards including the 2010 Sigcomm lifetime achievement award for contributions to the field of networking, the Usenix lifetime achievement award, and an honorary doctorate from KTH (Royal Institute of Technology, Sweden). She holds about 100 patents and a Ph.D. from MIT in computer science.
upcoming events view all
At the Cutting Edge: The Next Disruptors of Sustainable Innovation (panel)
25 January 2017 | San Francisco, CA
Conferences & Talks
Applications of AI and Machine Learning in Healthcare: Focus on Comorbidities (Co-occurrence of Multiple Chronic Conditions)
1 March 2017 | London, England
Conferences & Talks